Modifiable Areal Unit Problem

Often the area (or regional) level rather than the individual level is the target of inference in geographic and environmental science studies. The use of aggregated data sources while performing geographic (or spatially) related inquires may well lead the researcher into an encounter with the modifiable areal unit problem (MAUP). Geographic analysis requires one to not only consider geographic area—the scale—but also the resolution of the data to be analyzed. The relationships between variables will change as the scale of measurement (resolution) changes. The relative size, scale, level of detail, or depth of penetration that characterizes an object or activity relates to the MAUP. This entry provides a cursory overview of the MAUP as it relates to geographic inquiry.

The outcome of two different geographic (or spatial) aggregations produce different degrees of within-area homogeneity; in geographic terms, the generic name for this is modifiable areal unit problem. The MAUP is common to all geographic (spatially) aggregated data. It consists of two interconnected parts. First, there is ambiguity about what constitutes the objects of spatial study—identified as the scale and aggregation problem. Second, there are the implications this holds for the methods of analysis commonly applied to aggregated data. The MAUP requires the researcher to ask the questions of how to incorporate scale into methods of geographic analysis, how to determine and identify the operational scale of geographic phenomenon, how relationships between variables change as the resolution (scale of measurement) increases or decreases, and to what extent information on geographic relationships at one scale can be used to make inferences about relationships at other scales.

The term modifiable is used because neither the choice of the number of geographic units (the scale of the analysis) nor their particular arrangement (how partitioning or zoning is selected given the scale of analysis) is essential and any number of other choices can be made. The MAUP consists of two problems, one statistical and the other geographical, and it is often difficult to isolate one from the other. Statistically, the underlying concern is that the analyst may not be using the appropriate statistic for the problem being addressed. Geographically, the problem is intrinsic to the observation of the effects of different geographic (or spatial) aggregations and the subsequent interpretation of the patterns revealed by the different aggregations.

Hence, it is imperative that the MAUP be analyzed in the context of a well-defined model that identifies the different scale components (from the individual level upward) underlying the behavior of each of the variables.